Today in Generative Media

China's Kling genAI video; Artists flee Instagram for Cara; Big Tech defends AI use

• China’s No 2 short video app Kuaishou unveils Sora-style product amid rush to catch up in AI (South China Morning Post) 13 more crazy examples and link to the project below 👇 (X)

• Artists are fleeing Instagram to keep their work out of Meta’s AI (Washington Post)

• Big Tech Launches Campaign to Defend AI Use (Hollywood Reporter)

• The state of generative AI in Hollywood: A special report (Variety)

• How A.I. Made Mark Zuckerberg Popular Again in Silicon Valley (New York Times)

• Adobe terms clarified: Will never own your work, or use it for AI training (9to5Mac)

• Generative AI Is The New Persuasion Edge (Forbes)

• How a photo of OpenAI’s Sam Altman, enhanced by AI, sparked a journalistic debate at GeekWire (GeekWire)

• OpenAI is ‘exploring’ how to responsibly generate AI porn (Unusual Whales)

• Cartwheel generates 3D animations from scratch to power up creators (TechCrunch)

• Exclusive interview with Raspberry Pi CEO: New $70 AI kit 'a watershed moment for us' (ZDNet)

• PlayCanvas adds 3DGS Support into Editor (RadianceFields)

• CVPR 2024 Paper Topics and Totals (Tableau)

• Why Does Diffusion Work Better than Auto-Regression? (YouTube)

• JAX implementation of VQVAE/VQGAN autoencoders (GitHub)

• DiLightNet: Fine-grained Lighting Control for Diffusion-based Image Generation (SIGGRAPH 2024, project page)

• Matching Anything By Segmenting Anything (CVPR 2024, project page)

• SF-V: Single Forward Video Generation Model (project page)

• Adversarial Generation of Hierarchical Gaussians for 3D Generative Model (project page)

• Physics3D: Learning Physical Properties of 3D Gaussians via Video Diffusion (project page)

• Dynamic 3D Gaussian Fields for Urban Areas (project page)

• pOps: Photo-Inspired Diffusion Operators (project page)

• Flash Diffusion Accelerating Any Conditional Diffusion Model for Few Steps Image Generation (project page)

• NeRF-Insert: Local 3D editing with multimodal control signals (project page)

• Step-aware Preference Optimization: Aligning Preference with Denoising Performance at Each Step (project page)

• MagicPose4D: Crafting Articulated Models with Appearance and Motion Control (project page)

• GenS: Generalizable Neural Surface Reconstruction from Multi-View Images (arXiv)

• Stealing Image-to-Image Translation Models With a Single Query (arXiv)

• LiveSpeech: Low-Latency Zero-shot Text-to-Speech via Autoregressive Modeling of Audio Discrete Codes (arXiv)

• Audio Mamba: Bidirectional State Space Model for Audio Representation Learning (arXiv)

• 3D-HGS: 3D Half-Gaussian Splatting (arXiv)

• 🎵 Stable Audio Open v1.0 is now available on 🥪 tost.ai 🥳 with

  RunPod’s serverless L40 🔥 (X)

• all-in-one comfyUI experience: create workflows and turn them into scalable APIs, side by side (X)

• Making non-player characters in video games actually playable thanks to AI (X)

• This is what I dreamed of doing as a kid (X)